GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
alvinchirchir
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
alvinchirchir
Model Tree
Qwen/Qwen2.5-0.5B-Instruct
Input Modalities
Output Modalities
Supported Functionality
Greedy decoding, same prompts as the base model:
| Prompt | Base Qwen2.5-0.5B-Instruct | pirate-speak-0.5b |
|---|---|---|
| What is the capital of France? | The capital of France is Paris. | 'Tis Paris, ye scallywag! But don't be too aggrree, or I'll loose my sail! Haul yer sails, and I'll take the decks! |
| What should I name my dog? | Choosing the right name for your dog is an important decision... | A fine mate, but don't worry about the details, or ye'll be drowned too! |
| How do I fix a flat tire? | To fix a flat tire, follow these steps: 1. Check the Tire Pressure... | A flat tire's the first thing to happen when you land on the ocean floor, but don't ye be too quick about it, or else ye'll just end up with a wet deck! |
python
from transformers import pipelinepirate = pipeline("text-generation", model="alvinchirchir/pirate-speak-0.5b")messages = [{"role": "user", "content": "How do I make a cup of coffee?"}]print(pirate(messages, max_new_tokens=80)[0]["generated_text"][-1]["content"])
Or with the model and tokenizer directly:
python
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizermodel_id = "alvinchirchir/pirate-speak-0.5b"tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained(model_id, dtype=torch.float32)inputs = tokenizer.apply_chat_template([{"role": "user", "content": "Any tips for a job interview?"}],add_generation_prompt=True, return_tensors="pt", return_dict=True,)output = model.generate(**inputs, max_new_tokens=80, do_sample=False)print(tokenizer.decode(output[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))
Training used Qwen's default chat template with no custom system message, so results are most consistent without one.
Intended use: learning and experimentation - a hands-on example of LoRA fine-tuning for style, and a fun chat toy.
Not intended for: anything where correct information matters.
Known limitations:
Data: 15 hand-written prompt/response pairs of everyday questions answered in pirate speak, formatted as conversational prompt/completion pairs so the loss is computed only on the assistant reply.
| Prompt | Response |
|---|---|
| What's the weather like today? | Arrr, the skies be grey as bilge water, matey! Best bring yer coat. |
| What is the capital of France? | 'Tis Paris, ye scallywag, a fine port for fancy landlubbers. |
| Tell me a joke. | Why don't pirates take a shower before they walk the plank? Because they'll just wash up on shore later! Har har! |
| How do I learn Python? | Start with the basics, matey, then practice like ye be sailin' rough seas - a little each day! |
| What is machine learning? | 'Tis teachin' a contraption to spot patterns, like readin' the stars to find treasure. |
Hyperparameters:
| Setting | Value |
|---|---|
| LoRA rank / alpha / dropout | 16 / 32 / 0.05 |
| LoRA target modules | all linear layers (attention + MLP) |
| Trainable parameters | 8.8M (1.75%) |
| Epochs / steps | 6 / 48 |
| Batch size | 2 |
| Learning rate | 2e-4 (linear decay) |
| Max grad norm | 1.0 |
| Precision | float32 (no mixed precision) |
| Final training loss |
Hardware: Apple M1 Pro (PyTorch MPS backend).
Software: Transformers 4.57.6, TRL 1.13.0, PEFT 0.19.1, PyTorch 2.9.1.
After training, the LoRA adapter was merged into the base weights with merge_and_unload(), so the model loads like any regular Transformers model with no PEFT dependency.
This model is a modified version of Qwen2.5-0.5B-Instruct, Copyright 2024 Alibaba Cloud, licensed under the Apache License 2.0. The base model is available at https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct.
Modifications: LoRA fine-tuning on 15 pirate-speak conversation examples, with the adapter merged into the model weights. The architecture, tokenizer, and chat template are unchanged.
This derivative is distributed under the same Apache License 2.0; a copy is included in this repository as LICENSE.
If you use the base model in research, please cite the Qwen team (citation from the Qwen2.5-0.5B-Instruct model card):
bibtex
@misc{qwen2.5,title = {Qwen2.5: A Party of Foundation Models},url = {https://qwenlm.github.io/blog/qwen2.5/},author = {Qwen Team},month = {September},year = {2024}}@article{qwen2,title={Qwen2 Technical Report},author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},journal={arXiv preprint arXiv:2407.10671},year={2024}}
| ~0.74 |